Cost per point of capability

How much you pay per point of measured capability. This is a division, not an opinion: output price divided by the Artificial Analysis intelligence index. Both numbers are shown so you can redo the calculation yourself.

Where the scores come from. The capability indices are produced by Artificial Analysis, relayed through OpenRouter's catalogue API. They are not our measurement, and we do not run benchmarks ourselves. We record them as a level 4 source and show them next to the price we observed. Prices are what the reseller charges.

Model	Intelligence	Coding	Output price	Cost per point
inclusionAI: Ling 3.0 Flash	20.1	50.6	$0.063	$0.00313433
Z.ai: GLM 5.3 Flash	27.8	71.5	$0.0923	$0.00332014
inclusionAI: Ling 3.0 Flash Fin	22.6	55.6	$0.1232	$0.00545133
inclusionAI: Ling 3.1 Flash	41.1	n/a	$0.225	$0.00547445
DeepSeek: DeepSeek V4.1 Flash (batch)	24.7	n/a	$0.150	$0.00607287
Z.ai: GLM 5.3 Flash (batch)	27.8	71.5	$0.180	$0.00647482
Anthropic: Claude Haiku 5.5 (batch)	36.2	n/a	$0.250	$0.00690608
inclusionAI: Ling 3.0 Flash VL	24.6	57.0	$0.180	$0.00731707
DeepSeek: DeepSeek V4 Flash 0423	24.4	56.2	$0.182	$0.00745902
Google: Gemma 4 26B A4B	16.7	39.3	$0.127	$0.00760479
Upstage: Solar Mini 4	24.1	n/a	$0.200	$0.00829876
DeepSeek: DeepSeek V4 Flash 0731	34.3	69.1	$0.300	$0.00874636
OpenAI: gpt-oss-120b	11.6	30.4	$0.104	$0.00896552
OpenAI: GPT-6 Luna (batch)	27.8	n/a	$0.250	$0.00899281
DeepSeek: DeepSeek V4 Flash 0731 (batch)	34.3	69.1	$0.330	$0.00962099
Xiaomi: MiMo-V2.6-Flash	26.7	n/a	$0.280	$0.01048689
OpenAI: gpt-oss-20b (batch)	10.0	20.7	$0.112	$0.0112
Qwen: Qwen3.5-9B (batch)	21.8	28.7	$0.250	$0.01146789
Meta: Llama 3.1 8B Instruct	6.9	5.4	$0.080	$0.01159420
OpenAI: gpt-oss-120b (batch)	11.6	30.4	$0.136	$0.01172414
Mistral: Mistral Small 3	6.7	n/a	$0.080	$0.01194030
Inception: Mercury 2.5	12.3	n/a	$0.150	$0.01219512
OpenAI: gpt-oss-20b	10.0	20.7	$0.130	$0.013
Mistral: Ministral 3 8B 2512 (batch)	5.5	9.7	$0.075	$0.01363636
Anthropic: Claude Haiku 5.5	36.2	n/a	$0.500	$0.01381215
OpenAI: GPT-5 Nano (batch)	13.0	n/a	$0.200	$0.01538462
OpenAI: GPT-5.6 Luna (batch)	37.3	71.4	$0.600	$0.01608579
StepFun: Step 3.5 Flash	17.0	n/a	$0.300	$0.01764706
DeepSeek: DeepSeek V3.2	21.5	44.2	$0.380	$0.01767442
OpenAI: GPT-6 Luna	27.8	n/a	$0.500	$0.01798561
DeepSeek: DeepSeek V4 Flash Vision Exp	34.8	65.0	$0.6468	$0.01858621
Qwen: Qwen3.5-9B	13.3	28.7	$0.250	$0.01879699
Google: Gemini 2.5 Flash Lite (batch)	10.4	n/a	$0.200	$0.01923077
DeepSeek: DeepSeek V4.1 Flash	24.7	n/a	$0.500	$0.02024291
MiniMax: MiniMax M3.1 Flash	28.9	n/a	$0.600	$0.02076125
Google: Gemma 3 4B	4.8	2.7	$0.100	$0.02083333
Mistral: Ministral 3 3B 2512	4.8	4.8	$0.100	$0.02083333
DeepSeek: DeepSeek V4.1 Flash 1017	28.5	n/a	$0.600	$0.02105263
Upstage: Solar Pro 4	28.2	52.7	$0.600	$0.02127660
IBM: Granite 4.0 Micro	5.1	n/a	$0.112	$0.02196078
Tencent: Hy3 preview	25.3	58.8	$0.600	$0.02371542
Amazon: Nova Micro 1.0	5.9	n/a	$0.140	$0.02372881
Microsoft: Phi 4	5.9	n/a	$0.140	$0.02372881
Nex AGI: Nex-N2-Pro	41.7	59.1	$1.00	$0.02398082
DeepSeek: DeepSeek V3.2 Exp	16.6	n/a	$0.410	$0.02469880
OpenAI: GPT-4.1 Nano (batch)	7.8	11.1	$0.200	$0.02564103
Mistral: Mistral Small 4 (batch)	11.3	26.6	$0.300	$0.02654867
Z.ai: GLM 4.7 Flash	14.9	n/a	$0.400	$0.02684564
NVIDIA: Nemotron 3 Nano 30B A3B	8.9	14.4	$0.240	$0.02696629
Google: Gemma 4 31B	14.7	43.4	$0.400	$0.02721088
Mistral: Ministral 3 8B 2512	5.5	9.7	$0.150	$0.02727273
Xiaomi: MiMo-V2.6-Pro	29.9	n/a	$0.870	$0.02909699
IBM: Granite 4.2 8B	5.1	22.4	$0.150	$0.02941176
OpenAI: GPT-5.4 Nano (batch)	20.7	56.1	$0.625	$0.03019324
OpenAI: GPT-5 Nano	13.0	n/a	$0.400	$0.03076923
Reka Flash 3	6.4	n/a	$0.200	$0.03125
OpenAI: GPT-5.6 Luna	37.3	71.4	$1.20	$0.03217158
Mistral: Ministral 3 14B 2512	6.0	14.4	$0.200	$0.03333333
Qwen: Qwen3 VL 32B Instruct	11.9	n/a	$0.416	$0.03495798
Kwaipilot: KAT-Coder-Pro V2	33.7	59.5	$1.20	$0.03560831

60 models ranked, out of those for which we hold both a published price and a capability score. A model missing from this table is missing a figure, not judged.